Elsevier

Social Science & Medicine

Volume 55, Issue 12, December 2002, Pages 2243-2252
Social Science & Medicine

Race/ethnicity and depressive symptoms: a cross-cultural/ethnic comparison among university students in East Asia, North and South America

https://doi.org/10.1016/S0277-9536(02)00003-5Get rights and content

Abstract

To examine manifestations of depressive symptomatology among undergraduate students in East Asia, North and South America, responses to the Center for Epidemiologic Studies Depression Scale (CES-D) were compared across Japanese (n=310), Anglo-American (n=377), Native American (n=353), and Argentinean (n=110) undergraduate students. Japanese reported a significantly higher level of low positive affect, leading to significantly higher total CES-D scores, whereas their negative symptoms score was comparable to scores of Anglo-Americans and Argentineans. Although Native Americans were more likely to endorse negative symptoms, their low positive affect score was comparable to those of Argentineans. Argentineans appear to suffer less from depressive symptoms. Results from a Differential Item Functioning analysis, using Anglo-Americans as the reference group, indicated that: (1) the manifestation of depressive symptoms seemed to be similar for Anglo-Americans and Argentineans, except for low positive affect; (2) Native Americans tended to favor somatic symptoms over affective (depressive) symptoms; (3) responses to positive affect questions could possibly be biased not only for Japanese but also for people in North America; i.e., the expression of positive affect might be enhanced in North American culture, while inhibited in Japanese culture.

Introduction

Major depression is one of the most common diseases in industrialized countries and is further estimated to become the second most debilitating disease worldwide in 2020 (Murray & Lopez, 1996). The detection of depression in community settings, as well as prevention and treatment strategies, will increasingly become a major public health issue warranting heightened attention. Along with the World Mental Health 2000 Initiative, cross-cultural/ethnic research is of growing importance for health professionals. Cross-cultural/ethnic studies are critical because psychological impairment may be both expressed and influenced largely by local sociocultural environmental factors (e.g., Devins, Beiser, Dion, Pelletier, & Edwards, 1997; Robins, 1989; Vernon, Roberts, & Lee, 1982). Thus, specific response patterns and psychometric properties of assessment instruments across ethnic/cultural populations require further investigation.

The Center for Epidemiologic Studies Depression Scale (CES-D), developed in the US for the use in community surveys to identify groups “at high-risk” of depression in a general population (Radloff, 1977), is now widely used as a measure of depressive symptomatology for persons in Europe (e.g., Fava, 1983; Fuhrer & Rouillon, 1989; Soler et al., 1997), Asia (e.g., Cho & Kim, 1998; Iwata & Saito, 1987; Mackinnon, McCallum, Andrews, & Anderson, 1998) and elsewhere. Ideally, standard or widely accepted measurements, such as the CES-D, should be equivalent across race/ethnicity, cultures, and regions (Flaherty et al., 1988). That is, items of such scales should be free from item- and scale-level bias by exogenous variables such as gender, race/ethnicity, and culture. These are essential issues warranting heightened attention for both scientific research and health services because (1) most mental disorders and minor psychiatric impairment are defined by their manifestations as responded by a subject rather than direct disease markers, and (2) the racial/ethnic diversity of citizens has become more prevalent in industrial countries (cf. Kuo, 1984).

Some studies have addressed race/ethnicity-specific response patterns to items of the CES-D (e.g., Golding, Aneshensel, & Hough, 1991; Iwata, Okuyama, Kawakami, & Saito, 1989; Roberts, Vernon, & Rhoades, 1989; Vernon et al., 1982). For example, Japan–US comparison studies, using the CES-D for middle school students (Iwata, Saito, & Roberts, 1994) and adult workers (Iwata, Roberts, & Kawakami, 1995), revealed that the Japanese have a tendency to inhibit the expression of positive affect as measured by positively oriented (POS) items, while response patterns to negatively oriented (NEG) items were generally comparable. In addition, the authors (Iwata, Saito, & Roberts (1994), Iwata, Roberts, & Kawakami (1995)) suggested that not only the Japanese but some other Asian and racial/ethnic groups would have a similar response tendency. The corresponding results have been reported for Koreans in South Korea (Cho & Kim, 1998) and Korean immigrants in Canada (Noh, Kasper, & Chen, 1998). These results have been obtained by simple statistics such as an average frequency distribution, t-test and/or ANOVA. However, these analytic procedures may not be appropriate to examine the expression or inhibition of positive feelings because it is a “relativistic phenomenon” in comparison with negative feelings/symptoms. To extract more conclusive results, another approach should be utilized, such as an analysis of differential item functioning (DIF; Holland & Thayer, 1988).

The present study attempts to address the item-level bias, or DIF, of the CES-D using response data of four racial/ethnic groups from East Asia, South and North America. The DIF refers to a situation in which respondents who differ from one another on a certain exogenous variable (e.g., race/ethnicity), but who resemble one another on a certain latent continuum of interest (e.g., depressive symptoms), show different probabilities of endorsing an item intended to measure the latent continuum (Clauser & Mazor, 1998). It is important to clarify the difference between DIF and item impact. Item impact is a crude difference in response to an item by level of the exogenous variable. It may vary because respondents differ with respect to the latent variable. In contrast, DIF is expected to be the same across levels of the exogenous variable once conditioned on the latent variable (Cole, Kawachi, Maller, & Berkman, 2000). Taking these differences on a latent continuum into account when investigating response probability is a unique feature of DIF analysis.

While some studies have reported using psychiatric rating scales for Central American populations (e.g., Aguilar & Berganza, 1990; Vega, Kolody, Hough, & Figueroa, 1987) and Puerto Ricans (e.g., Vera et al., 1991), few studies have applied such scales to South American populations. Hence, there remains a limited understanding about the manifestation of psychiatric symptoms (depressive symptoms in particular) and expression style or response tendency among people in South America. In this study, we analyzed CES-D data collected in Argentina and compared manifestations of depressive symptomatology with other racial/ethnic groups in East Asia (Japanese) and North America (Anglo- and Native-Americans). Particular attention was paid to DIF of POS and NEG items across groups as an investigation of ethnocultural differences in the expression or inhibition of positive affect.

Section snippets

Survey sites and subjects

The sample characteristics of the four racial/ethnic groups are shown in Table 1. Subjects missing three or more of the 20 CES-D items were excluded from this study. To enhance the comparability of group comparisons, we restricted the study sample to undergraduate respondents between ages 18 and 24 years.

(1) Japanese data: Of the 317 Japanese students from three universities in Tokyo and Fukuoka, who had completed the Japanese version of the CES-D (Shima, Shikano, Kitamura, & Asai, 1985), 310

Mean rating scores on individual items, subscales, and total scale

The left four columns of Table 2 show the gender- and age-adjusted mean scores (with standard errors) of individual items, subscales, and total scale for the four racial/ethnic groups. The items are listed along with the traditional subscales (Radloff, 1977). Mean scores that are significantly higher or lower than those of Anglo-Americans are boldfaced or italicized, respectively.

For NEG items, Native Americans tended to show higher scores; their scores on six items (3 DEP, 2 SOM, 1 INT) were

Discussion

This study is the first to investigate the difference in endorsement level of depressive symptoms among four racial/ethnic groups from East Asia, North and South America. To our knowledge, data on depressive symptomatology from South American respondents have seldom been reported in international journals and are the first to be reported here. Although Iwata et al. (1995) compared endorsements of depressive symptoms between demographically matched Japanese and Anglo-American white-collar

Acknowledgements

This study was supported partly by the “Space Utilization Frontiers Joint Research Projects” promoted by the National Space Development Agency of Japan. We thank Dr. Cecilia M. Onaha, Graduate School for Advanced Studies (at the time of study), Kyoto, Japan, Dr. Eva Y. Deykin, former professor at Harvard School of Public Health, and the Data Access Committee of the National Center for American Indians and Alaska Natives Mental Health Research, University of Colorado Health Sciences Center, for

References (51)

  • C Chien et al.

    Depression in TaiwanEpidemiological survey utilizing CES-D

    Psychiatrica et Neurologica Japonica [Seishin-Shinkeigaku Zasshi]

    (1985)
  • M.J Cho et al.

    Use of the Center for Epidemiologic Studies Depression (CES-D) Scale in Korea

    Journal of Nervous and Mental Disease

    (1998)
  • B.E Clauser et al.

    Using statistical procedures to identify differential item functioning test items

    Educational Measurement: Issues and Practice

    (1998)
  • R.B Darlington

    Another look at “cultural fairness”

    Journal of Educational Measurement

    (1971)
  • M Davidson et al.

    A model of occupational stress

    Journal of Occupational Medicine

    (1981)
  • G.M Devins et al.

    Cross-cultural measurements of psychological well-beingThe psychometric equivalence of Cantonese, Vietnamese, and Laotian translations of the Affect Balance Scale

    American Journal of Public Health

    (1997)
  • G.A Fava

    Assessing depressive symptoms across culturesItalian validation of the CES-D self-rating scale

    Journal of Clinical Psychology

    (1983)
  • J.A Flaherty et al.

    Developing instruments for cross-cultural psychiatric research

    Journal of Nervous and Mental Disease

    (1988)
  • R Fuhrer et al.

    La version francaise de l’echelle CES-D (Center for Epidemiologic Studies-Depression Scale). Description et traduction de l’echelle d’autoevaluation [the French version of the CES-D (Center for Epidemiologic Studies-Depression Scale)]

    European Psychiatry

    (1989)
  • J.M Golding et al.

    Responses to depression scale items among Mexican-Americans and non-Hispanic whites

    Journal of Clinical Psychology

    (1991)
  • P.W Holland et al.

    Differential item performance and the Mantel–Haenszel procedure

  • J.J Hurrell et al.

    Exposure to job stressA new psychometric instrument

    Scandinavian Journal of Work and Environmental Health

    (1988)
  • N Iwata et al.

    Responses of Japanese and American university students to the STAI items that assess the presence or absence of anxiety

    Journal of Personality Assessment

    (2000)
  • N Iwata et al.

    Prevalence of depressive symptoms in a Japanese occupational settingA preliminary study

    American Journal of Public Health

    (1989)
  • N Iwata et al.

    Relationships of the Todai Health Index to the General Health Questionnaire and the Center for Epidemiologic Studies Depression Scale

    Japanese Journal of Hygiene

    (1987)
  • Cited by (140)

    View all citing articles on Scopus
    View full text